2013 Volume 18 Issue 2 Pages 151-156
Spatially explicit classification is an essential planning technique in managing natural forests having structural heterogeneity on a landscape scale. Operationally, forest planners demarcate classification boundaries on-site, based on their empirical knowledge. This study examined factors affecting spatially explicit classification decisions by expert planners in natural forest management in central Hokkaido, Japan. A case analysis was conducted at the University of Tokyo Hokkaido Forest to identify the criteria used for classification decision-making. We examined a total of 930 inventory plots within four major management types ('softwood selection harvest,' 'hardwood selection harvest,' 'pre-harvest,' and 'regeneration activity required') during 2005 to 2009. We calculated tree basal area (BA) and density, species composition, and diameter distribution in each management type. 'Softwood selection harvest' had high abundance and density of conifers (Abies sachalinensis in particular) at marketable size classes [diameter at breast height (DBH) ≥ 13 cm]. In 'hardwood selection harvest,' a large BA of broad-leaved trees (mainly Tilia japonica and Acer mono) was observed at middle- and large-size classes. 'Pre-harvest' had high densities of both coniferous and broad-leaved trees (mainly A. sachalinensis and Betula ermanii) of small diameter. The BA and tree density were relatively low in 'regeneration activities required.' Results from the classification and regression tree analysis indicated that the amount of marketable timber (25 cm ≤ DBH < 59 cm) had the primary effect on spatially explicit classification decisions in natural forest.
